计算机工程与应用 ›› 2012, Vol. 48 ›› Issue (3): 228-231.

• 工程与应用 • 上一篇    下一篇

应用马尔可夫随机场的金属疲劳断口条带分割

王 琳1,刘 君1,黎 明2   

  1. 1.南昌航空大学 信息工程学院,南昌 330063
    2.南昌航空大学 无损检测教育部重点实验室,南昌 330063
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2012-01-21 发布日期:2012-01-21

Striation segmentation of metal fatigue fracture based on MRF

WANG Lin1, LIU Jun1, LI Ming2   

  1. 1.School of Information Engineering, Nanchang Hangkong University, Nanchang 330063, China
    2.Key Laboratory of Nondestructive Test, Nanchang Hangkong University, Nanchang 330063, China
  • Received:1900-01-01 Revised:1900-01-01 Online:2012-01-21 Published:2012-01-21

摘要: 研究了金属疲劳断口图像的分割问题,提出了一种基于马尔可夫随机场(MRF)的金属疲劳断口图像的条带分割方法。由于疲劳断口图像中的纹理记录了整个断裂过程中的受力情况,通过对疲劳断口的条带纹理进行分析可以反演断裂的过程,因此研究疲劳断口图像的分割可以对失效分析有重要的科学价值。文中构造了图像的马尔可夫随机场模型,并且提出了一种基于该模型的图像分割算法。马尔可夫随机场模型是一种描述图像结构的概率模型,能够充分利用图像的空间相关信息,能够实现对低信噪比的金属疲劳断口图像进行条带分割。结果表明算法具有收敛速度快、稳健性好等优点。

关键词: 疲劳断口, 图像分割, 马尔可夫随机场, 参数估计

Abstract: This article has studied the fatigue of metal fracture image division question and a Markov Random Field(MRF)-based striation segmentation methord for metal fatigue fracture image is proposed. Because of the texture in the fatigue fracture image recording steess of the fracture process, through analyzing the striation textue of the fatigue fracture can invers the fracture process. Therefor, studying the fatigue fracture image segmentation has improtant scientific value for failure analysis. Markov Random Field model of image is structured and image segmentation algorithm is proposed in this article. Markov random field model is a probability model describing the structure of image. It makes full use of the image of space-related information and can be achieved with low signal to noise ratio of metal fatigue fracture striation segmentation. The results show that the algorithm has fast convergence and good stability.

Key words: fatigue fracture, image segmentation, Markov Random Field(MRF), parameters estimation